<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->
hsohn3/mayo-bert-uncased-wordlevel-block512-ep10
This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.3171
- Epoch: 9
Model description
- base_model: bert-base-uncased
- block_size: 512
- tokenizer: ehr-bert-wordlevel-uncased
Intended uses & limitations
More information needed
Training and evaluation data
- MAYO visit-level texts
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: float32
- mlm_probability: 0.15
- batch_size: 8
- epochs: 10
Training results
Train Loss | Epoch |
---|---|
3.0885 | 0 |
2.8340 | 1 |
2.7975 | 2 |
2.6720 | 3 |
2.4868 | 4 |
2.1750 | 5 |
1.8143 | 6 |
1.0948 | 7 |
0.4915 | 8 |
0.3171 | 9 |
Framework versions
- Transformers 4.20.1
- TensorFlow 2.8.2
- Datasets 2.3.2
- Tokenizers 0.12.1